Search papers, labs, and topics across Lattice.
EVA-Client is an open-source framework designed to streamline the deployment, data collection, and evaluation of manipulation policies on real robots, effectively bridging the gap between policy servers and physical hardware. Its architecture allows for modular integration of robot backends, inference strategies, and transport middlewares, enabling easy updates without affecting the entire system. Key results include a unified approach to real-time control and data collection that ensures each evaluation run contributes to the next training cycle, enhancing the efficiency of the policy iteration loop.
Each evaluation run in EVA-Client not only assesses performance but also enriches the training dataset, creating a continuous feedback loop for policy improvement.
We present EVA-Client, an open-source framework for deployment, data collection, and evaluation of trained manipulation policies on real robots. Sitting between a policy server and the physical hardware, EVA-Client unifies the real-robot stages of the policy iteration loop within a single codebase. It makes three contributions. First, a component-decoupled architecture in which robot backends, inference strategies, and transport middlewares form an orthogonal grid: adding a robot or a strategy touches only its own layer. Second, inspectable execution through Debug, Collect, and Eval workflows, with modes ranging from open-loop simulation to continuous real-time control. Third, every evaluation run doubles as a data collection, recording full rollouts in training-ready format alongside exhaustive logs and a side-by-side comparison viewer, so each evaluation feeds the next round of training rather than ending as an unrecorded impression. EVA-Client further consolidates major real-time inference strategies, synchronous and asynchronous execution, ACT-style temporal ensembling, Real-Time Chunking, and a naive-async ablation baseline, behind a single configuration surface.
